GROWTH 91 



results in more or less accelerated growth at the site of the wound. 

 In the case of a limb, the new growth is most rapid if removal is 

 effected at the level where autotomy (the spontaneous separation of 

 an appendage) occurs naturally; but it is not necessarily connected 

 with the property of autotomy, for there is regeneration if appen- 

 dages are cut through above or below this level, and there is regenera- 

 tion in appendages which cannot autotomize. 



The relation between regeneration and general growth and moult- 

 ing is puzzling. If a leg is removed from Blattella at a very early stage 

 in the moulting cycle, moulting is arrested until regeneration of the 

 limb is complete. It seems that the moulting hormone is not neces- 

 sary for regenerative growth. Indeed in the larva of Ephcstia (Ana- 

 gasta) regenerations of a wing rudiment can take place after the brain 

 and thoracic gland have been removed so that moulting hormone 

 cannot be secreted. 



The influence which moulds the outward form of the new appen- 

 dage seems to reside in the epidermis itself: regeneration is un- 

 affected by extirpation of the corresponding ganglion ; though when 

 this is done, neither muscles nor nerves are formed within the new 

 appendage. In some cases, regeneration is 'heteromorphous' : in 

 CarausiuSy if the antenna is cut through at the level of the flagellum, 

 it regenerates as an antenna; if it is amputated through one of the 

 two basal segments it commonly regenerates as a leg, complete with 

 tarsal claws. But though this tendency exists, there is no absolute 

 relation between the level of section and the type of regeneration. 



Thus, the phenomena of regeneration, like those of normal 

 growth, can be defined by certain laws - but they are laws of growth 

 and of regeneration; they cannot yet be expressed in the simpler 

 terms of physiology, far less in terms of physics and chemistry. 



Determination of sex 



The genetic theory of sex determination has been admirably des- 

 cribed by F. A. E. Crew in this series of monographs. This theory, 

 which is based very largely on studies of insects, will explain the 

 determination of sex both in ordinary sexual reproduction and in 

 parthenogenesis. It can probably be accepted that, in insects, both 

 primary and secondary sexual characters are determined by the 



